SOLVED FOG Multicast Starts Prematurely

  • I am running version 1.5.5 and each time I create a multicast group and schedule a task for those computers (36+/-) fog starts to image the first dozen or so without waiting for the others. The rest end up in a que and I end up forcing them to start. They then image in singlecast. I have extended the UDPMAXWAIT to 1000 but there is no change. I can’t find anything on this issue and I am not sure what logs to look at. Any help is appresciated.

  • @Sebastian-Roth This is exactly what it was. I feel a little embarrassed haha. Thank you for your help

  • Senior Developer

    @Gdelcimp Maybe I get this wrong but it sounds as if you start a normal group deploy instead of a group multicast deploy. I can imagine all of your computers doing single cast anyway. Do you see those 10-12 that start early go pretty much in sync? I mean do they all proceed in very much the same speed and on the same sector/byte count?

    Make sure you use this icon (or it’s equivalent in the Basic Tasks) to start a multicast session:

    If you are absolutely sure you created a multicast task then you might wanna take a look at the log file. Run tail -f /opt/fog/log/multicast.log in a command sehll on your FOG server before you schedule the session and then keep an eye on the log…

  • so to clarify, I am imaging 6 groups of computers, with each group having about 36 computers. Each group I register to fog, then when it asks if I want to image now I select no. I then create a group in fog, move the 36 computers that were just registered to that new group, schedule the task, then turn the 36 computers on. They then go through the process of getting to the multicast group but about 10-12 of them start almost immedialy, and the rest are put in a que for singlecast.

    Its interesting because the rest obviously see that they were scheduled to be imaged with the rest of the group, but somehow end up in singlecast.

  • @Sebastian-Roth You mean just try the same thing, but with a newly created group? I have not tried to create a new group and reimage the same computers, but each time image a group of computers I create a new group for those computers. I see the same thing happen with each group.

  • Senior Developer

    @Gdelcimp Have you ever tried to create a new group, add those hosts you want to image to that group and schedule a multicast deploy task for that group? Do you see the same thing happening when using a group multicast?